| 1 |
On Tue, 19 Dec 2017 20:11:04 -0600 |
| 2 |
R0b0t1 <r030t1@×××××.com> wrote: |
| 3 |
|
| 4 |
> I forgot most files were mirrored. So the infrastructure that is the |
| 5 |
> answer to my question is already in place. Consequently, I don't think |
| 6 |
> there's any reason to argue against this, unless it ultimately ends up |
| 7 |
> being a ton of work to package small files (which I can't comment on). |
| 8 |
|
| 9 |
Biggest downsides I see of pushing patches to distfiles mirrors is it |
| 10 |
greatly desynchronizes the state of tree, similar to how you'd get |
| 11 |
desynchronization if parts of gentoo.git were sharded into different |
| 12 |
repositories. |
| 13 |
|
| 14 |
Comprehending changes made downstream require additional steps and |
| 15 |
additional tooling. |
| 16 |
|
| 17 |
Correlating patch changes against ebuild changes is even more effort. |
| 18 |
|
| 19 |
The only upside is it makes it slightly harder to abuse patch-reuse and |
| 20 |
have unintentional retroactive patching to existing ebuilds without -r |
| 21 |
bumps. |
| 22 |
|
| 23 |
But ... that's a double edged sword if that sort of thing is |
| 24 |
occasionally useful and sane. |
| 25 |
|
| 26 |
I don't know what the solution is here, but I don't think either |
| 27 |
strategies of "discourage it" or "encourage it" are the ultimate way |
| 28 |
forward. |
| 29 |
|
| 30 |
Some other strategy must exist. But for now, sensible limits are an |
| 31 |
acceptable compromise. |
Archives